This invention relates to a novel peptide growth factor and, more particularly, to a heparin-binding growth factor derived from bovine uterus and human placenta.
In recent years a considerable number of growth factors derived from various animal cells have been isolated and characterized. Illustrative of these growth factors are nerve growth factor (NGF) which has been purified from several different cell sources, insulin-like growth factors (IGF-I and IGF-11), epidermal growth factor (EGF), fibroblast growth factor (FGF), platelet-derived growth factor (PDGF), endothelial cell growth factor (ECGF), somatomedins and transforming growth factors (TGF) derived from various tumors and virally transformed cells. For background information on these growth factors see, for example, the recent brief review articles by Kris et al., Biotechnology 3, 135-140 (1985); Dijk and Iwata, Ibid. 7, 793-798 (1989); and the comprehensive review in Hormonal Proteins and peptides, Ed. by Choh Hao Li, Vol. 12, xe2x80x9cGrowth Factors,xe2x80x9d Academic Press, 1984.
Although the existence of acidic and basic fibroblast growth factor (aFGF and bFGF) has been known for fifty years [Trowell, et al., J. Exp. Biol. 16, 60-70, (1939); Hoffman, Growth 4, 361-376 (1940)], only recently have they been purified and sequenced [Thomas et al., Proc. Natl. Acad. Sci. USA 81, 357-361 (1984); Gimenez-Gallego et al., Science 230, 1385-1388 (1985); Lemmon and Bradshaw, J. Cell. Biochem. 21, 195-208 (1983); Bohlen et al., Proc. Natl. Acad. Sci. USA 81, 5364-5368 (1984); and Esch et al., Proc. Natl. Acad. Sci. USA 82, 6507-6511 (1985)]. aFGF and bFGF have 55% sequence homology, suggesting that they arose by duplication and divergence from a common ancestral gene. Cells responding to the FGFs have between 103 and 105 high affinity receptors per cell [Neufeld and Gospodarowicz, J. Biol. Chem. 261, 5631-5637 (1986)]. The bFGF receptor has recently been purified, sequenced, and cloned [Lee et al., Science 245, 57-60 (1989)]. Both bFGF and aFGF compete for the same receptor and displace each other in radio-receptor assay [Neufeld, supra; Olwin and Hauschka, Biochemistry 25, 3487-3492 (1986)]; however, two forms of the receptor appear to have different affinities for aFGF and bFGF. Recently five additional proteins have been reported by cDNA cloning to have homology to the FGFs (hst, int-2, FGF-5, FGF-6, and KGF or FGF-7) [Dickson and Peters, Nature 326, 833 (1987); Yoshida et al., Proc. Natl. Acad. Sci. USA 84, 7305-7309 (1987); Delli Bovi et al., Cell 50, 729-737 (1987);.Zhan et al., Mol. Cell. Biol. 8, 3487-3495 (1988); Marcis et al., Oncogene 4, 335-340 (1989); and Finch et al., Science 245, 752-755 (1989)]. All five cDNA sequences encode signal peptides, therefore suggesting that these five proteins are presumably secreted, unlike aFGF or bFGF which lack signal peptides.
In accordance with the resent invention a novel heparin-binding growth factor has been isolated from bovine uterus and human placenta. This novel heparin-binding growth factor, herein also designated as HBGF-8, is a 18.9 kDa polypeptide with a unique 25 N-terminal amino acid sequence as follows:
Gly-Lys-Lys-Glu-Lys-Pro-Glu-Lys-Lys-Val-Lys-Lys-Ser-Asp-Cys-Gly-Glu-Trp-Gln-Trp-Ser-Val-Cys-Val-Pro.
This novel growth factor binds tightly to cation exchange resins and to Heparin-Sepharose(copyright) and is stable to acetone precipitation and labile in acid.
HBGF-B was as active as acidic fibroblast growth factor (aFGF) and slightly less active than bFGF in the mouse NIH 3T3 fibroblast mitogenic assay system with an intrinsic specific activity of 5000 dpm/ng under standard assay conditions.
In an illustrative example and based upon total activity in the acetone extracts of bovine uterus stimulating 3H-thymidine incorporation into DNA of serum-starved NIH 3T3 cells, a 6940 fold purification was achieved with an overall yield of HBFG-8 activity of 0.4%, using extraction of acetone powders and chromatographic separations at neutral pH. Approximately 18 xcexcg protein was obtained from 1.2 kg wet weight of tissue. HBGF-8 was clearly separated from 17.5 kDa bovine uterus basic fibroblast growth factor (bFGF) by purification and its N-terminal amino acid sequence analysed. The unique sequence as set forth above was found.
In accordance with another aspect of the invention, the complete coding sequence of CDNA clones representing the full size bovine HBGF-8 and human HBGF-8 have been developed.
Thus, the bovine HBGF-8 cDNA sequence was isolated from bovine uterus cDNA library using the above bovine HBGF-8 N-terminal sequence clone as a probe. The cDNA sequence contains 1196 nucleotides and encodes a 18.9 kDa (Mr=18,902) protein of 168 amino acids, including a 32 amino acid leader sequence. The cDNA coding sequence starts at nucleotide position 170 to 673.
The human HBGF-8 cDNA sequence was isolated from human placenta cDNA library using bovine HBGF-8 cDNA fragment as a probe. The cDNA sequence contains 995 nucleotides and encodes a 18.9 kDa (Mr=18,942) protein of 168 amino acids, including a 32 amino acid lead sequence. The cDNA coding sequence starts at nucleotide position 252 to 755.
Comparison of the bovine and human HBGF-8 showed that HBGF-8 is highly conserved. Of 168 amino acids, 163 were identical. Differences occur at amino acid positions 3, 4 and 7 in the leader sequence and at amino acid positions 130 and 147 in the mature protein of the complete 168 amino acid sequence.
The human and bovine cDNA protein sequences and the human and bovine cDNA sequences for HBGF-8 are as follows:
Met Gln Ala Gln Gln Tyr Gln Gln Gln Arg Arg Lys Phe Ala Ala 15
Ala Phe Leu Ala Phe Ile Phe Ile Leu Ala Ala Val Asp Thr Ala 30
Glu Ala Gly Lys Lys Glu Lys Pro Glu Lys Lys Val Lys Lys Ser 45
Asp Cys Gly Glu Trp Gln Trp Ser Val Cys Val Pro Thr Ser Gly 60
Asp Cys Gly Leu Gly Thr Arg Glu Gly Thr Arg Thr Gly Ala Glu 75
Cys Lys Gln Thr Met Lys Thr Gln Arg Cys Lys Ile Pro Cys Asn 90
Trp Lys Lys Gln Phe Gly Ala Glu Cys Lys Tyr Gln Phe Gln Ala 105
Trp Gly Glu Cys Asp Leu Asn Thr Ala Leu Lys Thr Arg Thr Gly 120
Ser Leu Lys Arg Ala Leu His Asn Ala Glu Cys Gln Lys Thr Val 135
Thr Ile Ser Lys Pro Cys Gly Lys Leu Thr Lys Pro Lys Pro Gln 150
Ala Glu Ser Lys Lys Lys Lys Lys Glu Gly Lys Lys Gln Glu Lys 165
Met Leu Asp 168
Met Gln Thr Pro Gln Tyr Leu Gln Gln Arg Arg Lys Phe Ala Ala 15
Ala Phe Leu Ala Phe Ile Phe Ile Leu Ala Ala Val Asp Thr Ala 30
Glu Ala Gly Lys Lys Glu Lys Pro Glu Lys Lys Val Lys Lys Ser 45
Asp Cys Gly Glu Trp Gln Trp Ser Val Cys Val Pro Thr Ser Gly 60
Asp Cys Gly Leu Gly Thr Arg Glu Gly Thr Arg Thr Gly Ala Glu 75
Cys Lys Gln Thr Met Lys Thr Gln Arg Cys Lys Ile Pro Cys Asn 90
Trp Lys Lys Gln Phe Gly Ala Glu Cys Lys Tyr Gln Phe Gln Ala 105
Trp Gly Glu Cys Asp Leu Asn Thr Ala Leu Lys Thr Arg Thr Gly 120
Ser Leu Lys Arg Ala Leu His Asn Ala Asp Cys Gln Lys Thr Val 135
Thr Ile Ser Lys Pro Cys Gly Lys Leu Thr Lys Ser Lys Pro Gln 150
Ala Glu Ser Lys Lys Lys Lys Lys Glu Gly Lys Lys Gln Glu Lys 165
Met Leu Asp 168
In the bovine HBGF-8 protein sequence, the following amino acid replacements of the human sequence are found:
Ala-3xe2x86x92Thr; Gln-4xe2x86x92Pro; Gln-7xe2x86x92Leu; Glu-130xe2x86x92Asp; and Pro-147xe2x86x92Ser.